International organisation for standardisation organisation internationale de normalisation

4.2Assignment of output documents & editors

• 
  Editors
  

Documents	Editors
RVC Core Experiments	Kazuo-san

• 
  Output documents
  

No.	Title	P.A.	Editing period
	Description of Core Experiments in RVC	N	4 weeks

4.3Review of input contributions

• 
  On-going issues
  

Doc. No.	Category	Title	Authors
m16332	MPEG-B Video	An efficient dataflow design for implementing MPEG-4 AVC decoder in the RVC framework	Jérôme Gorin, Mickaël Raulet
	Notes	“a new way to test the efficiency and the reliability of the CAL tools, especially Synthesizing tools, by modelling one of the most complex decoder provided by MPEG” Deblocking filter needs to be checked. Bitstream parser needs verified. What is described in VTL FDIS is explained in detail as a report. [Recommendation] Offline meeting (Tuesday Afternoon) to solve the deblocking filter problem.
m16450	MPEG-C Video	MPEG-4 ASP Decoder Description and a Case Study of the Decoder Design Process in the RVC framework	Hyungyu Kim, Sowon Kim, Minsoo Park Hwa Seon Shin, Byeongho Choi, Chungku Yie, Euee S. Jang
	Notes	ECMAScript-based ASP parser description is proposed. Design process including conformance testing is proposed. [Recommendation] Include the proposed design process into PDAM1 (Study)
m16482	MPEG-C Video	Status and Future Plan of Video Tool Library (VTL)	Hwa Seon Shin, Sung Moon Chun, Hyungyu Kim, Byeongho Choi, Euee S. Jang
	Notes	Status of the current VTL descriptions of FU is given. [Recommendation] Start a CE on Token definition (HS, Christophe, Mikael, Chris)
m16470	All General/All	FU network of inverse scan, inverse quantization and inverse transform in AVC High Profile	Gwo Giun (Chris) Lee, He-Yuan Lin, Jia-Wei Liang, Ming-Jiun Wang
	Notes	[Recommendation] conduct crosschecking (HYU) and accept it in VTL (WD).
m16333		FU parameterization and FU code generation	M. Raulet, M. Wiplize and J. W. Janneck
	Notes	A report on duplication and parameterization of FUs [Recommendation] Start a CE on the reusability of token and parameterization

• 
  Core Experiment on Encoding Tools
  

Doc. No.	Category	Title	Authors
m16398	MPEG-C Video	AVC Entropy Coding and Bitstream Generation for the MPEG RVC Encoding Tools	Hussein Aman-Allah, Ihab Amer, Marco Mattavelli
	Notes	Outline of AVC Entropy coding modules in high level. Status report [Recommendation] the encoding tools should demonstrate the efficiency and usefulness in a generic way to be applicable to many applications. [Recommendation] update CE description to specify the objectives of the work. (Marco)
m16409	MPEG-C Video	Advances in the CE: Development of RVC Encoding Tools	Ihab Amer, Marco Mattavelli
	Notes	Status report
m16413	MPEG-C Video	AVC Intra Prediction, Transform and Quantization for the MPEG RVC Encoding Tools	Karim Maarouf, Ihab Amer, Marco Mattavelli
	Notes	Outline of AVC intra prediction, transform and quantization is given. Status report
m16431	MPEG-C Video	AVC Inter-frame Prediction for the MPEG RVC Encoding Tools	Ehab Asaad Hanna, Ihab Amer, Marco Mattavelli
	Notes	Status report

• 
  RVC for Graphics
  

Doc. No.	Category	Title	Authors
m16403	MPEG-C Video	A study on RVC based Graphics codec	Seungwook Lee, Bonki Koo, Kyoungsoo Son, Daiyong Kim, Mingxiao Chen, Euee S. Jang

4.4Output Document Processing

4.4.1CE description update (Editor: Kazuo-san)

• 
  CE 3 (RVC Encoding tools) update
  
  • 
    Update on the objective of this CE will be integrated.
    
• 
  CE 4 (RVC Conformance Testing)
  
  • 
    Existing two codecs (MPEG-4 SP, MPEG-4 AVC ConstrainedBP) will be checked first during the conformance testing CE till the next meeting.
    
• 
  CE 5 (Reusability of tokens and parameterization)
  
  • 
    Update of the CE description discussed and updated
    
• 
  CE 6 (RVC-BSDL extensions)
  
  • 
    Update of the CE description discussed and updated
    

4.5RVC-related schedule of the 88^th meeting

Day	Topic	Room	Time	Status
Monday	Video Plenary (planning of this week)	Video	14 - 15	DONE
	RVC time allocation	Wilcox	16 -1615	DONE
	Review of input contributions (on-going issues)	Wilcox	1615 – 1930	DONE
	RVC output decision	Wilcox	1730 - 18	DONE
Tuesday	RVC Conformance	Wilcox	930 – 12	DONE
	Joint meeting with 3DG on RGC	3DG (Pioneer)	14 - 15	DONE
	Core Experiments (Encoding Tools)	Wilcox	15 – 17	DONE
	RVC Conformance (cont’d)	Wilcox	17 – 1720	DONE
	Review of Input contribution (M16333)	Wilcox	1720 - 1640	DONE
Wednesday	CE update review	Wilcox	1430 – 1530	DONE
	RVC Conformance (cont’d)	Wilcox	1530 – 17	DONE
	RVC Vision	Wilcox	17 - 18	DONE
Thursday	CE update review	Wilcox	13 – 1530	DONE
Friday	Video Plenary	Video		DONE

Output Document:

No.	Title	TBP	Available
10551	Description of Core Experiments in RVC	N	09/04/24

5Explorations – 3D Video

The goal of 3D video is to generate interpolated views from available videos of multiview camera configurations. The target application is mostly seen for upcoming generations of various (auto-) stereoscopic displays, either requiring multiple views internally or providing means for baseline adjustment. In the new format, only a low number (1) of video sequences shall be transmitted, but rendering of additional views would be enabled by associated depth information.

In the exploration experiments preceding the Maui meeting, further progress was made in improving (automatic) depth estimation and view rendering / interpolation. Two approaches were investigated, namely layered depth video (including background video plus depth) and stereo/multiview plus depth, both of which seem to be conceptually working and suitable for producing anchors in an upcoming CfP. However, according to the subjective viewing judgements that were again performed in Maui, deficiencies in synthesis quality are mainly caused by false depth estimation.
Following the Call for test material and depth maps that had been issued in Lausanne, responses were received which in particular provide hand-generated depth maps either for sequences from the existing test set, or along with new sequences (2). Even though more effrt seems to be necessary in this, satisfactory view synthesis results are now achieved for the cases of 8 sequences in total (previously only 3), including cases of high depth variation and extremely localized content for which automatic algorithms typically fail. As also the depth maps look more smooth and natural, it is concluded that for the first time it is reasonable to perform the second step in anchor preparation for a CfP, which is coding experiments, using AVC/MVC for both video and depth map coding. Main achievements in this activity have been review of technical developments from the exploration experiments, further clarification about the vision, applications and requirements, and planning of next steps. The planned exploration experiments are therefore solely designed for the purpose of CfP preparation:

• 
  EE1: Depth map generation; goal: Improving semi-automatic methods, get good/better depth maps for LDV rendering and multi-view based rendering
  
• 
  EE4: Coding experiments; goal: Find best data rate points (video, depth) for upcoming CfP
  

If the results are satisfying, a draft CfP could be issued in July. To further progress on this, a document “Evaluation and Testing of 3D Video Coding” (N10649) was produced, which

• 
  Describes viewing methodologies for stereoscopic and autostereoscopic displays, and an expert viewing method
  
• 
  Has the goal to get a "measurable" video quality on a more granular scale (not binary or ternary decision as in the viewing performed so far for the purpose of evaluating the performance of depth estimation and view synthesis algorithms).
  

Further, the document “Applications and Requirements of 3D Video Coding” was slightly updated (N10570).

Documents reviewed in AHG (see AHG report)

m16295	Revisions of Applications & Requirements on 3D Video Coding	Anthony Vetro
m16317	Results of Exploration Experiments in 3D Video Coding for Dog Data Set	Yongzhe Wang Philipp Merkle Karsten Müller
m16318	3DTV Exploration Experiments on Pantomime data set	Ivana Radulovic Per Fröjdh
m16319	MPEG 3DV EEs on Leaving_Laptop	Po-Lin Lai Dong Tian Patrick Lopez Paul Kerbiriou
m16326	3DV results on Dog sequence	Carmen CHENG Yan HUO Yu LIU
m16328	Results of Exploration Experiments in 3D Video Coding, described in w10360, for Alt Moabit sequence.	Olgierd Stankiewicz Krzysztof Wegner Krzysztof Klimaszewski
m16356	Report of 3D/FTV Exploration Experiment with Champagne Tower	Takanori Senoh Kenji Yamamoto Ryutaro Oi Tomoyuki Mishina Makoto Okui
m16357	Proposal of Depth Map Estimation Method in Response to Call for 3D Test Material: Depth Maps & Supplementary Information	Takanori Senoh Kenji Yamamoto Ryutaro Oi Tomoyuki Mishina Makoto Okui
m16388	3DV EE1 & EE2 Results on Newspaper sequence	Seok Lee Jaejoon Lee Ilsoon Lim Jin Young Lee Ho-Cheon Wey Du-Sik Park
m16389	Error-resilient Free-viewpoint Image Generation for FTV	Masayuki Tanimoto Toshiaki Fujii Mehrdad Panahpour Tehrani Hisayoshi Furihata Menno Wildeboer
m16390	Depth Estimation Reference Software (DERS) 3.0	Masayuki Tanimoto Toshiaki Fujii Mehrdad Panahpour Tehrani Kazuyoshi Suzuki Menno Wildeboer
m16391	Semi-automatic Depth Estimation for FTV	Masayuki Tanimoto Toshiaki Fujii Mehrdad Panahpour Tehrani Norishige Fukushima Kazuyoshi Suzuki Menno Wildeboer
m16394	EE1: Results of Depth Estimation on 'Pantomime? Sequence	Cheon Lee Yo-Sung Ho
m16395	EE2: Results of View Synthesis on 'Pantomime? Sequence	Cheon Lee Jiho Park Yo-Sung Ho1
m16396	3-D Test Sequence - Multiview Video and Depth Map	Eun-Kyung Lee Yo-Sung Ho
m16400	3DV EE3 on Champagne_tower sequences	Yin Zhao Lu Yu
m16405	LDV Reference Software for View Synthesis	Yin Zhao Lu Yu Fons Bruls Lincoln Lobo
m16406	3DV/FTV EE results of depth estimation and view synthesis on "lovebird1" sequence	Gun Bang Gi Mun Um Namho Hur Jinwoong Kim
m16407	PSPNR measurement for evaluating view synthesis quality	Yin Zhao Lu Yu
m16411	Depth Estimation algorithm in SADERS1.0	Gun Bang Jaeho Lee Namho Hur Jinwoong Kim
m16417	Improving view synthesis results based on depth quality measure	Jaewon Sung Byeong-Moon Jeon
m16418	3DV EE results on Newspaper sequence	Jaewon Sung Byeong-Moon Jeon
m16419	Philips response to new Call for 3DV Test Material: Arrive book & Mobile	fons bruls rene klein gunnewiek patrick van de walle
m16420	Philips & Zhejiang Uni response to new Call for 3DV Test Material: Champagne Tower	fons bruls rene klein gunnewiek patrick van de walle Yin Zhao Lu Yu
m16421	Philips (in coop with 3D4YOU) response to new Call for 3DV Test Material: Beergarden	fons bruls rene klein gunnewiek patrick van de walle rene van de vleuten
m16422	Philips 1st 3DV synthesis results using new test material for Arrive book, Mobile, Beergarden & Champagne Tower	fons bruls rene klein gunnewiek patrick van de walle
m16423	Philips 3DV EE results	fons bruls
m16432	Results of Exploration Experiments in 3D Video for Lovebird2	Sehoon Yea Zafer Arican Anthony Vetro
m16460	Additional results of Exploration Experiments in 3D Video Coding, described in w10360, for Alt Moabit sequence.	Olgierd Stankiewicz Krzysztof Wegner Krzysztof Klimaszewski
m16471	3DV/FTV EE Report on Doorflower sequence	Shinya Shimizu Hideaki Kimata

Output documents:

No.	Title	TBP	Available
10570	Applications and Requirements of 3D Video Coding	N	09/04/24
10552	Description of Exploration Experiments in 3D Video Coding	N	09/04/24
10649	Evaluation and Testing of 3D Video Coding	N	09/04/24